heemod
heemod is an R package that helps researchers build and analyze Markov models for health economic evaluations. It supports various types of models and analyses to assess the cost-effectiveness of healthcare interventions.
At a glance
Use when
Conducting health economic evaluations using state-transition models, especially when time dependency or advanced sensitivity analyses are needed.
Avoid when
When users lack R programming skills or when the clinical problem requires non-Markov modeling approaches such as microsimulation or agent-based models.
Inputs
Transition probabilities, cost and utility values per state, time horizons, discount rates, patient characteristics, and parameter distributions for sensitivity analysis.
Outputs
Cost-effectiveness results (e.g., ICERs), sensitivity analysis outputs (e.g., tornado diagrams, PSA results), subgroup analyses, and model trajectories over time.
How it works
heemod implements Markov models, including semi-Markov and non-homogeneous Markov models, with support for deterministic and probabilistic sensitivity analyses, heterogeneity analysis, and time dependency on both state-time and model-time. It aligns with established health economic modeling standards from Briggs et al. (2011) and Siebert et al. (2012), enabling comprehensive model-based reporting for health technology assessment.
- HTA domains
- Clinical Effectiveness, Costs & Economic Evaluation, Organisational aspects
- Categories
- Cost-effectiveness ModellingModel ValidationReporting Standards & Best PracticeTransparency
- Assumptions
- Model structure accurately represents disease progression; transition probabilities and other parameters are appropriately estimated and time dependency is correctly specified.
- Strengths
- Flexible modeling of time dependencies, supports probabilistic and deterministic sensitivity analyses, integrates with R for reproducible research, follows established HTA guidelines.
- Limitations
- Requires proficiency in R programming; may be complex for beginners; limited support for non-Markovian structures like discrete event simulation.
- Also known as
- heemod R package
Questions this answers
- › What is the cost-effectiveness of a healthcare intervention over time?
- › How do results vary under different assumptions or patient characteristics?
- › How does timing of events affect outcomes in a model?
- › What is the impact of uncertainty in model parameters?
- › How do outcomes differ across patient subgroups?
- › How can Markov models be structured and reported according to best practices?
References & sources
Similar by meaning
Beta record. Generated from the primary source via AI extraction and independent audit, pending final human review.